Releasesby Glyph2026.02
We're excited to announce major updates to the Gradium API and Gradium Studio, delivering enhanced text-to-speech (TTS) and speech-to-text (STT) capabilities for developers and enterprises.
Also shipped Aug 16, 2026Gradium in August 2026 →
Feature policy rules (General availability)
With this release, we are pleased to announce the general availability of feature policy rules. A feature policy can carry a YAML body that conditionally blocks object creation based on attributes of the request. For example, you can permit tables in general but block temporary tables, or permit tasks but block serverless (no-warehouse) tasks.
Govern Replit at scale
Replit has released a new Admin API, Audit Logs, and Workspace Settings to support enterprise AI adoption. The update includes comprehensive audit logs covering over 50 events with native streaming to SIEM systems.
OpenAI Agents SDK 0.21.1
- feat(core): add model call timeouts feat(sandbox): add run-scoped sandbox working directories feat(sandbox): allow Docker sandboxes to disable networking feat(sandbox): add Modal sandbox resource options fix(core): honor exact call approval decisions fix(core): count Responses requests without usage fix(core): close all MultiProvider children after...
More from GradiumEvery Gradium release →
- Model selector in Gradium Studio : the Text-to-Speech page now has a Model dropdown to switch between the...
- Model selector in Gradium Studio : the Text-to-Speech page now has a Model dropdown to switch between the production model and gradium-tts-beta , showing each model_name with a copy button. Shareable studio links : pre-configure a session with ?voice= , ?model_name= , ?text= and ?language= and share the full setup as a single link.
Speech-to-Text requests now require a language in json_config
Speech-to-Text requests now require a language in json_config . Our ASR gives its best results when the language is known, so set it whenever you can: en , fr , es , pt , or de . When the language is not known ahead of time, pass any and the model detects it from the audio. Requests that already set a language or any are unaffected.
Our live translation models ( stt-translate for translated transcription and s2s-translate for...
Our live translation models ( stt-translate for translated transcription and s2s-translate for Speech-to-Speech) now produce more natural, more accurate translations, with better context understanding across a conversation. The quality gains are most noticeable on conversational speech. See Speech-to-Speech and the Speech-to-Text overview to get started.
A new Gradium Text-to-Speech model is live and is now the default
A new Gradium Text-to-Speech model is live and is now the default. Existing voices, including custom voices, work as-is. The model reads the structured entities that break voice agents in production, such as phone numbers, email addresses, IBANs and reference codes, with no pre-processing or text normalization needed.
A new Gradium Text-to-Speech model is available in public beta
A new Gradium Text-to-Speech model is available in public beta. It handles complex text natively, with no pre-processing or normalization on your side: phone numbers in national and international formats, email addresses, IBAN numbers, time expressions, reference codes, measurements, URLs and ordinals.
A new Speech-to-Text model is now live on the standard endpoint
A new Speech-to-Text model is now live on the standard endpoint. Transcription is more robust on low volume audio, so soft speakers and far field microphones are recognised with improved reliability. Endpointing is more reactive, so the end of an utterance is detected sooner and voice agents can start responding with less delay.
Create a voice from a written description, no audio sample needed
Create a voice from a written description, no audio sample needed. Describe the accent, age and character you need, generate up to five candidates, and keep your favourite. The voice works with the same Text-to-Speech endpoint as any catalogue voice. In our blind listening evaluation, native speakers judged how well voices matched accent prompts.